Download Cholinergic and Adrenergic Tones in the Control of Heart Rate

Comp. Biochem. Physiol. Vol. 118A, No. 1, pp. 131–139, 1997
Copyright  1997 Elsevier Science Inc.
ISSN 0300-9629/97/$17.00
PII S0300-9629(96)00402-1
Cholinergic and Adrenergic Tones
in the Control of Heart Rate in Teleosts.
How Should They be Calculated?
Jordi Altimiras,1 Abbas Aissaoui,1 Lluis Tort,1 and Michael Axelsson2
1
Unitat de Fisiologia Animal, Facultat de Ciències, Universitat Autònoma de Barcelona,
E-08193 Bellaterra, Spain; and 2 Department of Zoophysiology, Göteborg University, Medicinaregatan 18,
S-413 90 Göteborg, Sweden
ABSTRACT. Cholinergic and adrenergic tones were calculated for three different teleost fish species: Gadus
morhua, Labrus bergylta, and Sparus aurata using atropine as a muscarinic receptor antagonist and either sotalol
or propranolol as β-adrenoceptor antagonists. Depending on the order of administration of atropine and the
two β-adrenoceptor antagonists, it was observed that propranolol but not sotalol enhanced cholinergic tone.
Thus, if propranolol is used to determine autonomic cardiac influences, it has to be injected after atropine and
not before. Differences in intrinsic heart rate were observed between treatments in two of the three species
studied, suggesting the activity of a non-cholinergic non-adrenergic mechanism in heart rate control in fish.
Different models to calculate cholinergic and adrenergic tones are discussed. The additive model described by
other authors is appropriate provided that no interaction exists between cholinergic and adrenergic influences.
We demonstrate no interaction in the species studied in this experiment. Finally, a modification of the additive
model that uses R-R interval instead of heart rate in the computation is proposed. This is justified with a computer
simulation in terms of the linearity of the response given the reciprocal relationship between R-R interval and
heart rate. comp biochem physiol 118A;1:131–139, 1997.  1997 Elsevier Science Inc.
KEY WORDS. Adrenergic tone, atropine, autonomic interaction, cholinergic tone, heart rate control, propranolol, sotalol, teleost
INTRODUCTION
The autonomic nervous system is the main short-term modulator of heart rate in vertebrates. A double antagonistic
innervation has been demonstrated in all vertebrate groups
with exception of cyclostomes, elasmobranchs, dipnoans
and some groups of teleosts. Even in species lacking a direct
spinal autonomic innervation of the heart the circulating
catecholamines released from the chromaffin tissue will
have a positive chronotropic and inotropic effect on the
heart (7,17).
The role of vagal efferents in inducing heart rate (HR)
changes was first demonstrated using electrical stimulation
(13,18,36). More contemporary studies have used specific
agonists/antagonists to further study the double control of
the heart (8,11). Muscarinic receptor antagonists induce a
rise in heart rate by blocking cholinergic post-synaptic receptors and reducing the inhibitory parasympathetic drive.
On the other hand, β-adrenoceptor antagonists antagonize
Address reprint requests to: Dr. Jordi Altimiras, Department of Biological
Sciences, University of Aarhus, Building 131, Universitetsparken, DK8000 Aarhus C, Denmark. Tel. 45-89-422589; Fax 45-86-194186; E-mail:
[email protected].
Received 31 May 1996; accepted 2 October 1996.
competitively the natural catecholamines at the level of
cardiac β-receptors, which results in a negative chronotropic and inotropic effect.
Several models of autonomic chronotropic cardiac control have been described. Although these models were based
on experiments using electrical stimulation in mammalian
hearts (35,37), they have been widely employed to characterize cholinergic and adrenergic tonus on the heart in most
vertebrate groups including teleosts (2,8,11). They adopted
the mathematical formulation reported by Lin and Horvath
(29), which is based in the additive model of Rosenblueth
and Simeone (35).
In the light of new findings in relation with autonomic
cardiac control in mammals, it seemed important to re-evaluate and adapt the methods to attain an accurate quantification of cholinergic and adrenergic tones. First question
is, how should the cholinergic and adrenergic tones be calculated? Second, and most important, the additive model
of Rosenblueth and Simeone (35) does not account for possible interactions between the cholinergic and adrenergic
autonomic influences. The existence of such interaction in
resting conditions has been termed ‘‘accentuated antagonism’’ (28) and is still controversial in mammals (27). To
our knowledge, it has not been described in teleosts.
J. Altimiras et al.
132
This topic can be approached pharmacologically by the
injection of autonomic antagonists in different order. Axelsson et al. (5) used this protocol in Myoxocephalus scorpius
and did not find differences. If interaction exists one would
expect that, for example, after the administration of a βadrenoceptor antagonist, there would be a compensatory response of the parasympathetic branch in order to minimize
the effect of the reduced adrenergic influence on the heart.
Such ‘‘compensatory demand’’ has been observed in mongrel dogs in exercise conditions (9).
on the lid of the chamber in which the fish was maintained.
With this setup, the stainless steel tube worked as the reference electrode being in direct contact with the water, and
minimum interferences were introduced in the signal.
The ECG signal was amplified and filtered in a Grass
Polygraph Recorder System (Model 7D) and digitalized in
a PC computer equipped with a LabPC1 card and LabView
software (National Instruments Co.). Once digitalized,
the QRS complex for each beat was detected in real time
and instantaneous heart rate stored.
MATERIALS AND METHODS
Species and Handling Procedures
Pharmacological Administration
and Experimental Protocol
The study involved several experiments in different teleost
species: 11 individuals of the Atlantic cod Gadus morhua
(558 6 174 g), 10 individuals of the ballan wrasse Labrus
bergylta (172 6 17 g), and 22 individuals of the Mediterranean sea bream Sparus aurata (210 6 30 g) were used in
the study. The animals were kept in well aerated tanks for
at least 1 week before any surgical procedure took place.
G. morhua were maintained at 10°C, L. bergylta at 20°C
(the animals were acclimated at that temperature for at least
1 week) and S. aurata at 16°C.
For comparison, results from a previously published study
in the bull-rout M. scorpius (5) will also be used in the present work. In this early study, the same experimental protocol described here was employed (see below).
Two different anesthetics were used. G. morhua was anaesthetized with MS222 (0.1g ⋅ L21, SIGMA). 2-Phenoxyethanol (0.75 mL ⋅ L21, SIGMA) was used for S. aurata
and L. bergylta. The anesthetic dose was chosen in order to
get a similar induction time (2 min) for both agents. Once
anaesthetized, the fish was weighted to the nearest gram and
placed ventral side up on an operation table, and the gills
continuously irrigated during the entire surgery.
For the drug injections, the ventral aorta was chronically
cannulated using a PE–50 tubing via the afferent branchial
artery of the third gill arch (5). The cannula was filled with
heparinized saline (100 I.U. ⋅ mL21). The cannula was secured with two skin sutures. The surgery was performed in
less than 20 min and the animals were returned to the experimental chamber, where they were allowed to recover
for at least 24 hr.
After the recovery period, heart rate was recorded for 1
hr before any manipulation of the animals took place. After
this initial recording period, the animals received a first injection of either a muscarinic receptor antagonist (Atropine
sulfate, 1.2 mg ⋅ kg21, SIGMA) or a β-adrenergic antagonist
(S-propranolol hydrochloride, 3.0 mg ⋅ kg21, FLUKA in L.
bergylta and S. aurata or Sotalol hydrochloride, 2.7 mg ⋅
kg21, Bristol Myers Squibb in G. morhua and S. aurata).
Thirty min after this injection the ECG was recorded for
another hour. Once the effects of the first injection were
recorded, the second antagonist (depending on which one
was given first) was injected and similarly, a 1 hr recording
was taken 30 min later (see Fig. 1 for details).
The same convention used in a previous paper (1) will be
employed to refer to the different treatments: GI for those
animals receiving the β-adrenergic antagonist (sotalol or
Heart Rate Measurements
Heart rate was obtained from the electrocardiographic signal. Two enamel-coated stainless steel wires (0.2 mm in diameter, supplied by Driver Harris S.A. and insulated by Aismalibar S.A.) were attached to the fish forming a bipolar
lead. Both electrodes were anchored to the skeleton surrounding the pericardial cavity: a first electrode midventrally where both cleithra bones join and a second electrode
between the fused structures of the pelvic girdle [see (19)
for a general anatomical description]. The depth at which
each electrode was inserted was established after anatomical
dissection of the different species. Both wires were sutured
to the skin in two places: ventrally between the pelvic fins
and dorsally at the base of the dorsal fin, preventing relative
displacements of the electrodes and improving the signal
quality. The wire electrodes were passed through a stainless
steel tube and soldered to an appropriate connector placed
FIG. 1. Schematic diagram showing the terminology for the
different treatments with the order of injection and its expected effects on the R-R interval.
Cholinergic and Adrenergic Tones in Teleosts
propranolol) before atropine and GII for those receiving atropine first.
Calculation of Cholinergic and Adrenergic Tones
We propose the use of the following equations for the quantification of cholinergic and adrenergic tones:
GI (Sotalol first)
Chol(%) 5
(R-R)β 2 (R-R)0
* 100
(R-R)0
Adr(%) 5
(R-R)β 2 (R-R)cont
*100
(R-R)0
GII (Atropine first)
Chol(%) 5
Adr(%) 5
(R-R)cont 2 (R-R)musc
* 100
(R-R)0
(R-R)0 2 (R-R)musc
*100
(R-R)0
where:
(R-R)cont —Control R-R interval.
(R-R)β —R-R interval after β-adrenoceptor blockade.
(R-R)musc —R-R interval after muscarinic receptor blockade.
(R-R)0 —R-R interval after complete autonomic blockade.
Chol(%)—Cholinergic tone in percent.
Adr(%)—Adrenergic tone in percent.
The rationale to use these equations will be discussed in
another section.
Statistics
The non-parametric Mann-Whitney test was used to assess
differences in intrinsic HR between GI and GII groups for
each species. An ANCOVA design with previous logarithmic transformation of the data was employed to test the
differences in cholinergic and adrenergic tone between GI
and GII using control HR as a covariant. All tests were
applied using SPSS software.
RESULTS
Heart rates and cholinergic and adrenergic tones in the different species depending on the injection order are shown
in Table 1.
This data set was statistically analyzed using an ANCOVA design in which control R-R interval was used as a
covariant. This design was preferred because it allows the
decomposition of intra-group variance in real inter-animal
variance and variance due to a different initial vagal drive.
In other words, part of the intra-group variance is reflecting
the linear relationship between vagal drive and control
133
R-R interval (see Fig. 2) and can be dismissed from the analysis using the covariant. No correlation was found between
control R-R interval and adrenergic tone.
The ANCOVA revealed differences ( p , 0.05) between
GI and GII in those experimental groups in which propranolol was used as a β-antagonist, with the exception of the
adrenergic tone for L. bergylta. No differences between GI
and GII were found when sotalol was used as a β-antagonist.
A comparison between propranolol and sotalol treatments
is shown for S. aurata in Fig. 3.
Finally, the intrinsic R-R interval (R-R)0, i.e., the R-R
interval after complete blockade with atropine and sotalol/
propranolol, differs between GI and GII in three of the four
species studied: L bergylta, M. scorpius, and S. aurata (Fig.
4). When (R-R)0 differs between treatments, the largest
R-R interval, i.e., the lowest HR, appears when atropine is
injected first in those species in which cholinergic tone is
more important than adrenergic tone (L. bergylta and S.
aurata) and the opposite is true for M. scorpius, in which
adrenergic tone is quantitatively more important than cholinergic tone.
DISCUSSION
Critique of Method
Atropine at the dose employed (1.2 mg ⋅ kg21 ) has a fast
antagonistic action on muscarinic receptors. In contrast
with the mild hypertension reported in humans (24,27), atropine does not induce changes in blood pressure in the
species studied [(5,8); Altimiras, unpublished observations].
Being so, there is no need to compensate for reflex changes
in heart rate associated with hypertensive effects (24). The
effects are also long lasting, and preliminary studies showed
no substantial recovery in vagal tone after 24 hr in S. aurata
or L. bergylta.
Beta-adrenergic antagonists are known to be less specific
and this needs to be taken into account when used to analyze adrenergic cardiac tone. Propranolol, for instance, is a
lipophilic molecule with membrane stabilizing properties
that also acts as a local anesthetic, independent of its effects
as a β-blocker (30). Aside from that, it enhances vagal tone
in humans (27) and the same effect has been observed for
atenolol (14) and sotalol (21). From this study, it appears
that propranolol but not sotalol facilitated vagal drive at
the doses employed in S. aurata, and this facilitation explains the differences in tones between GI and GII when
propranolol was used (Table 2). When propranolol is injected first (GI) it not only abolishes adrenergic tone but
it also enhances vagal drive, as can be seen by the larger
change in R-R interval induced by atropine when injected
after propranolol in comparison when injected before (see
Fig. 3). Interestingly, the same difference between GI and
GII treatments using propranolol was found in L. bergylta.
We also observed that the effects of β-antagonists disappeared more rapidly than the effects of atropine. At the
J. Altimiras et al.
134
TABLE 1. Heart rates before and after autonomic blockade
HRcont
G. morhua
L. bergylta
S. aurata
Sotalol
Propranolol
M. scorpius
HRatro
HRb
HR0
GI
GII
Only GII
Only GI
GI
40.0 6 3.7
90.4 6 9.2
37.2 6 2.4
84.5 6 8.6
45.8 6 1.3
111.6 6 3
33.5 6 2.2
72.9 6 7.2
38.3 6 1.9
106.7 6 4.1
*
39.8 6 0.8
88.3 6 3.4
63.4 6 5.4
66.1 6 2.2
48.3 6 2
89.0 6 2.2
83.3 6 3.9
53.0 6 1.5
60.4 6 4.0
43.9 6 2.2
32.3 6 1.1
83.2 6 2.2
79.9 6 4.6
36.7 6 1.9
*
*
*
73.8 6 3.6
72.2 6 1.5
42.3 6 1.3
66.4 6 4.4
57.5 6 2.5
41.8 6 1.4
*
*
GII
Data (%) are expressed as Mean 6 SEM.
*Indicates differences betweeen treatments (p , 0.05).
doses used, however, the effects on heart rate were consistently maintained for at least 5 hr after injection in preliminary studies.
Given the observed secondary effects of propranolol, especially its vagal accentuation, we suggest sotalol to be a
better β-adrenoceptor antagonist than propranolol, although the amiodarone-like activity of sotalol as an antiarrhythmic may also have undesirable interfering effects, not
yet described.
Finally, it is important to emphasize that adrenergic and
not sympathetic tones are obtained with the protocol used.
FIG. 3. R-R intervals in Sparus aurata in control conditions
and after partial and complete autonomic blockade. The
drugs employed in each case are specified: Atr—atropine,
Sot—sotalol, Pro—propranolol. Data plotted as Mean 6
SEM. GI treatment indicated as closed symbols, GII indicated as open symbols.
Sympathetic tone refers to the tonic activity of neural sympathetic innervation and, as such, does not include the effect of circulating catecholamines. Bretylium, a drug that
prevents the release of catecholamines from post-synaptic
adrenergic neurons, can be used to assess the adrenergic nervous tone as opposed to the total adrenergic tone (2). However, the severe and long lasting side effects of bretylium
and its difficult application in fish has prevented its exten-
FIG. 4. Intrinsic heart rate after complete pharmacologic
FIG. 2. Relationship between control R-R interval (of each
individual animal) and cholinergic tone (squares) and adrenergic tone (triangles) in Labrus bergylta. Notice the clear
increase in cholinergic tone when R-R intervals are longer
(and heart rates are reciprocally smaller). Each data point
corresponds to a single animal.
blockade in the different species using sotalol (top panel) or
propranolol (bottom panel) as a b-adrenoceptor antagonist.
Data from Myoxocephalus scorpius come from a previous
study (5). GI indicated as closed symbols, GII indicated as
open symbols. *—significant differences between treatments at p , 0.05.
Cholinergic and Adrenergic Tones in Teleosts
135
TABLE 2. Cholinergic and adrenergic tones in the different species in both treatments GI and GII
Cholinergic tone (%)
Adrenergic tone (%)
GI
G. morhua
L. bergylta
S. aurata
Sotalol
Propranolol
M. scorpius
15.1 6 2.9
50.7 6 11.6
40.7 6 8.2
84.1 6 13.7
13.2 6 3.3
*
*
GII
GI
GII
21.3 6 5.1
30.4 6 11.5
17.3 6 3.9
28.4 6 4.9
12.9 6 1.4
20.8 6 2.9
38.8 6 14.8
22.4 6 1.8
8.2 6 1.7
12.7 6 2.4
44.1 6 5.2
25.5 6 3.8
*
16.8 6 4.8
12.8 6 3.3
20.2 6 0.9
Data (%) are expressed as Mean 6 SEM.
*Indicates differences between treatments (p , 0.05).
sive use and β-adrenoceptor antagonists are still used as the
most common method to characterize the adrenergic tone
of the heart.
A second critique that could be raised lies on experimental design itself. A random administration of both treatments (GI and GII) in the same individuals in successive
days would have been the optimal experimental choice. The
long lasting effects of atropine, the unknown wash-out time
of the β-blockers and the difficulty of keeping cannulae
working for long-enough periods of time, prevented such
design.
Calculation of Tones
The previous literature in fish (2,5,11) consistently used the
equations given by Lin and Horvath (29) to calculate cholinergic and adrenergic tones (2). The equations we used,
as described in the Materials and Methods section, substitute HR values for R-R intervals.
In order to analyze in detail the differences between tones
calculated with either formulation, we simulated different
cholinergic and adrenergic influences at different control
HR using both treatments (GI and GII) in a computer
model. The output of this model for a GII treatment is
shown in Fig. 5A and B. We obtained two important conclusions:
1. Because tones are calculated as % changes, both calculations are independent of control HR, i.e., for a given
change in HR, no matter what the control HR was, we
obtain the same tonus. This property indicates that both
formulations can be used to compare different species
regardless of their control HR.
2. The two methods do not provide similar values at identical conditions, i.e., it is not the same to calculate tones
using HR as using R-R interval. This can be seen in Fig.
5A and B. Each plot shows cholinergic (X-axis) and adrenergic tones (Y-axis) calculated from R-R values versus
cholinergic tone or adrenergic tone (Fig. 5A and B, respectively) calculated from HR values. It is clear that
the two methods provide different results and only when
these tones are smaller than 20%, a close agreement is
obtained, as shown by the black areas in Fig. 5A and B.
Katona et al. (24) compared resting sympathetic and parasympathetic influences on sinoatrial function in athletes
and non-athletes quantifying HR changes after atropine or
propranolol injections. Their results confirmed the ‘‘accentuated antagonism’’ phenomenon previously described on
electrical stimulation preparations. However, ‘‘accentuated
antagonism’’ has been refuted by a more recent study that
used a similar protocol but analyzed R-R interval data rather
than HR (27). Kollai et al. recalculated the data obtained
by Katona et al. in terms of R-R interval, and showed that
there was no evidence for any ‘‘accentuated antagonism’’
when using R-R intervals instead of HR.
Despite these differences, the question of which is the
best model to use still remains. The rationale to use R-R
interval based calculations is that neural frequency discharge rates have been shown to be linearly related with
the beat-to-beat interval (26), and consequently it will be
reciprocally related with HR given the reciprocal relation
between R-R interval and HR (HR(Hz) 5 1/R-R(sec), see
Fig. 5C). Although these measurements were made in a
mammalian species, we think that it is fair to extend this
relationship to all vertebrate species. In conclusion, the
R-R interval-based method to quantify cholinergic and adrenergic tones in teleosts (and in other vertebrate groups)
offers a better estimate of these tones because it behaves
linearly in relation with the changes in R-R interval that
are, in turn, linearly related with nerve discharge frequency.
To further corroborate this conclusion we used our model
to calculate the cholinergic and the adrenergic tone in a
GII treatment if, at a given R-R interval (2 sec, equivalent
to a HR 5 30 bpm), the antagonists induced progressive
changes identical in magnitude but in opposite direction on
the R-R interval. The output for cholinergic tone is given
in Fig. 5D (it is identical for adrenergic tone because equal
influences were employed in the simulation). From the figure it is again clear that variations in R-R interval are linearly related with cholinergic tone when the R-R intervalbased model was used. Cholinergic tones obtained from the
J. Altimiras et al.
136
FIG. 5. A and B) Cholinergic (X-axis) and adrenergic tones (Y-axis) calculated from R-R values versus cholinergic (A) and
adrenergic tones (B) calculated from HR values. The shaded areas in each plot indicate a close agreement (5%) between the
tones calculated using HR and R-R values. C) Graphical plot of the reciprocal relationship between HR and R-R interval. D)
Effect of the use of different models (R-R based or HR-based models) on the calculation of cholinergic tone with stepwise
changes in R-R interval after muscarine receptor blockade. See text for more details.
HR-based model were not linear and consistently overestimated cholinergic tone when this was larger than 20%.
A survey of the fish literature provided cholinergic and
adrenergic tones in 15 species, being all the measurements
based on the HR based model and under a GII protocol.
Table 3 provides a recalculation of these tones using our
R-R based model. As described above (see also Fig. 5A and
B), there is a close agreement between both models for
those species with low tones such as Gadus morhua, Labrus
myxtus, Labrus bergylta and Myoxocephalus scorpius. However, large differences are found for those species with large
tones such as the antarctic fishes Pagothenia bernacchii and
Pagothenia borchgrevinkii and the goldfish Carassius auratus.
Interactions in the Autonomic Nervous System
As previously mentioned, both formulations discussed
above are based on an additive model of autonomic heart
rate control (35,37) that assumes no autonomic interaction.
The simplest formulation of this model is the one given by
Rosenblueth and Simeone (23):
HRc 5 c ∗ s ∗ HR0
where HRc is the heart rate, HR0 the intrinsic heart rate
and c and s the factors by which cholinergic and adrenergic
factors modify HR0, respectively.
In order to introduce an interaction between both, Cavero et al. (12) introduced a third multiplicative factor (i):
HRc 5 c ∗ s ∗ HR0 ∗ i
Interestingly, when Cavero et al. employed this second
model to their pharmacological data, they observed that
the interaction factor was not significantly different from 1
(i 5 0.985 p . 0.05), suggesting that there was no interaction between the cholinergic and adrenergic influences.
For the G. morhua data set we obtained an interaction
factor of 0.988, again suggesting no interaction between adrenergic and cholinergic outflows. Unfortunately, differences in intrinsic R-R interval between GI and GII treatments precluded the calculation of the interaction factor in
the other data sets.
Thus, it is concluded that an additive model of auto-
Cholinergic and Adrenergic Tones in Teleosts
137
TABLE 3. Survey of mean cholinergic and adrenergic tones for different fish species and recalculation of tones based on the
R-R model proposed in this paper
HR based
R-R based
Species
T (°C)
Chol %
Adr %
Chol %
Adr %
Ref.
Oncorhynchus kisutch
Gadus morhua
Hemitripterus americanus
Pagothenia borchgrevinki
Pagothenia bernacchii
Carassius auratus
Pollachius pollachius
Labrus mixtus
Labrus bergylta
Ciliata mustela
Raniceps raninus
Zoarces viviparus
Myoxocephalus scorpius
Thunnus thunnus
Katsuwonus pelamis
11–13
10–11
10–12
0–0.5
0–0.5
20–25
11–12
11–12
11–12
11–12
11–12
11–12
11–12
24–26
24–26
55.2
37.7
34.8
55.4
80.4
66.0
19.7
14.2
33.9
14.5
12.5
18.9
11.5
—
—
60.0
21.0
39.6
3.2
27.5
22.0
33.2
14.7
15.4
29.6
28.8
67.1
25.6
—
—
34
39
23
112
130
97
13
12
36
9
8
7
8
58.1
130.8
38
17
29
4
22
18
25
14
14
23
22
40
21
4.1
5.8
6
2
4
3
3
11
5
5
5
5
5
5
5
25
25
nomic cardiac control is appropriate to explain the adrenergic and cholinergic influences on heart rate in teleosts in
basal conditions.
However, autonomic interaction may still occur in other
conditions. In mammals neuropeptide Y is co-released from
post-synaptic noradrenergic fibers during exercise (32) and
attenuates vagal drive via inhibition of acetylcholine release
from preganglionar vagal neurones. In the dogfish, Squalus
acanthias, Xiang et al. (38) proposed an enhanced contractility mediated by neuropeptide Y due to a preferential stimulation of α-adrenoceptors on the heart. Neuropeptide Y has
also been found in the heart of two species of skates, Raja
erinacea and Raja radiata (10), but the functional significance is unclear. There is also one report of P12 purinoreceptors in the heart of the dogfish mediating a negative inotropic and chronotropic effects in the atrium (31). VIP has
been shown to be co-localized with acetylcholine in postganglionic parasympathetic neurons in dogs and rats (16)
and is responsible for a vagally induced tachycardia (20).
Intraarterially injected VIP produce an increase in the
stroke volume of the heart in the Atlantic cod, G. morhua,
but again the exact mechanism is not known (22).
Non-adrenergic Non-cholinergic Control
and Intrinsic Heart Rate
It is commonly accepted that neural autonomic influences
and circulating catecholamines are the main effectors on
the intrinsic pacemaker rhythm of the heart although there
are other mechanisms like the stretch of the pacemaker cells
and non-catecholamine hormonal effects that could also
play a role (15). According to this statement, the quantification of the cholinergic and the adrenergic tones of the
heart would explain most of the extrinsic effectors of cardiac
rhythm.
However, we observed a clear difference in intrinsic heart
rate between treatments in most of the species studied in
these experiments with the exception of G. morhua (Fig.
4). This result was completely unexpected because the only
difference between both groups was the order in which the
drugs were injected.
This observation could be related with the ‘‘excess tachycardia’’ reported in dogs (34) that is due to the positive
chronotropic effect of VIP (33). In fact, bilaterally vagotomized individuals of S. aurata have an intrinsic heart rate
of 62.0 6 2.7 bpm (Aissaoui, unpublished results), much
lower than the intrinsic rate with any of the pharmacological treatments reported here (83.2 6 2.2 and 79.9 6 4.6
bpm in GI, 79.9 6 4.6 and 72.2 6 1.5 bpm in GII, with
the sotalol and propranolol treatments, respectively).
The conflict of not knowing which is the actual intrinsic
heart rate at a given temperature does not compromise the
methodology proposed in this paper for assessing cholinergic
and adrenergic tones but raises a concern on how important
other effector mechanisms can be in determining the heart
rate of an animal at some given conditions.
We speculate that non-adrenergic/non-cholinergic
(NANC) effectors will be more important when the animals face environmental challenges (e.g., temperature, exercise, hypoxia). In this study, L. bergylta acclimated at
20°C (above the optimal temperature for the species),
showed the largest difference in HR0 between treatments
(106.7 6 4.1 bpm in GI and 88.3 6 3.4 bpm in GII).
Furthermore, in Pollachius pollachius there is an exerciseinduced tachycardia not eliminated by atropine and sotalol
in (5).
In summary, the results reported here could support the
existence of a NANC effector mechanism although further
work measuring stroke volume simultaneously with heart
rate would be required in order to exclude intrinsic mecha-
J. Altimiras et al.
138
nisms depending on venous return and estimate the contribution of such NANC mechanism to heart rate control in
fish.
We acknowledge the technical assistance of the people that took care
of the animals used in these experiments: Mr. Marc Puigcerver in the
aquarium of the Universitat Autònoma of Barcelona, the staff of the
Kristineberg Marine Research Station and Mrs. Barbro Blomgren in
the Animal Care Facility of the Department of Zoophysiology in Göteborg. J. A. and L. T. also want to thank Prof. Jarl-Ove Strömberg
for providing good working facilities for our experiments in Fiskebäckskil
(Sweden). J. A. was a recipient of a pre-doctoral grant from the Ministerio de Educación y Ciencia (Spain). L. T. received a travel grant
from the Dirección General de Relaciones Culturales y Cientı́ficas of
the Ministerio de Asuntos Exteriores (Spain). The work was financially
supported by a project grant (DGCICYT PB92-0637) to L. T. and
from the Swedish Forestry and Agricultural Research Council to
M. A.
14.
15.
16.
17.
18.
19.
20.
References
1. Altimiras, J.; Aissaoui, A.; Tort, L. Is the short-term modulation of heart rate in teleost fish physiologically significant? Assessment by spectral analysis techniques. Brazilian J. Med.
Biol. Res. 28(11–12):1197–1206;1995.
2. Axelsson, M. The importance of nervous and humoral mechanisms in the control of cardiac performance in the Atlantic
cod Gadus morhua at rest and during non-exhaustive exercise.
J. Exp. Biol. 137:287 –303;1988.
3. Axelsson, M.; Davison, W.; Forster, M.E.; Farrell, A.P. Cardiovascular responses of the red-blooded antarctic fishes Pagothenia bernacchii and P. borchgrevinki. J. Exp. Biol. 167:179–
201;1992.
4. Axelsson, M.; Driedzic, W.R.; Farrell, A.P.; Nilsson, S. Regulation of cardiac output and gut flow in the sea raven, Hemitripterus americanus. Fish Physiol. Biochem. 6(5):315–326;
1989.
5. Axelsson, M. Ehrenström, F.; Nilsson, S. Cholinergic and adrenergic influence on the teleost heart in vivo. Exp. Biol. 46:
179–186;1987.
6. Axelsson, M.; Farrell, A.P. Coronary blood flow in vivo in
the coho salmon (Oncorhynchus kisutch). Am. J. Physiol. 264:
R963–R971;1993.
7. Axelsson, M.; Farrell, A.P.; Nilsson, S. Effects of hypoxia and
drugs on the cardiovascular dynamics of the Atlantic hagfish
Myxine glutinosa. J. Exp. Biol. 151:297–316;1990.
8. Axelsson, M.; Nilsson, S. Blood pressure control during exercise in the Atlantic cod, Gadus morhua. J. Exp. Biol. 126:225–
236;1986.
9. Billman, G.E.; Dujardin, J.-P. Dynamic changes in cardiac vagal tone as measured by time-series analysis. Am. J. Physiol.
258:H896–H902;1990.
10. Bjenning, C.; Driedzic, W.; Holmgren, S. Neuropeptide Y-like
immunoreactivity in the cardiovascular nerve plexus of the
elasmobranchs Raja erinacea and Raja radiata. Cell Tissue Res.
255:481 –486;1989.
11. Cameron, J.S. Autonomic nervous tone and regulation of
heart rate in the goldfish, Carassius auratus. Comp. Biochem.
Physiol. 63C:341–349;1979.
12. Cavero, I.; Riggenbach, H.; Wall, M.; Gerold, M. Analysis of
cardiac chronotropic responses to some autonomic blocking
agents in conscious trained dogs. Eur. J. Pharmacol. 39:193–
202;1976.
13. Cobb, J.L.S.; Santer, R.M. Electrophysiology of cardiac func-
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
tion in teleosts: cholinergically mediated inhibition and rebound excitation. J. Physiol. 230:561–573;1973.
Cook, J.R.; Bigger, J.T., Jr.; Kleiger, R.E.; Fleiss, J.L.;
Steinman, R.C.; Rolnitzky, L.M. Effect of atenolol and diltiazem on heart rate period variability in normal persons. J. Am.
Coll. Cardiol. 17:480–484;1991.
Farrell, A.P. From hagfish to tuna: A perspective on cardiac
function in fish. Physiol. Zool. 64(5):1137–1164;1991.
Forssmann, W.G.; Triepel, J.; Daffner, C.; Heym, C.; Cuevas,
P.; Noble, M.I.M.; Yanaihara, N. Vasoactive intestinal peptide in the heart. Ann. N.Y. Acad. Sci. 527:405–420;
1988.
Fritsche, R.; Axelsson, M.; Franklin, C.E.; Grigg, G.G.; Holmgren, S.; Nilsson, S. Respiratory and cardiovascular responses
to hypoxia in the Australian lungfish. Respir. Physiol. 94(2):
173–187;1993.
Gannon, B.J. A study of the dual innervation of teleost heart
by a field stimulation technique. Comp. Gen. Pharmacol. 2:
175–183;1971.
Harper, W. Anatomy of fishes. Stuttgart: E. Schweizerbart’sche Verlagsbuchhandlung; 1975.
Hill, M.R.S.; Wallick, D.W.; Mongeon, L.R.; Martin, P.J.;
Levy, M.N. Vasoactive intestinal polypeptide antagonists attenuate vagally induced tachycardia in the anesthetized dog.
Am. J. Physiol. 269:H1467–H1472;1995.
Hohnloser, S.H.; Kligenheben, T.; Zabel, M.; Just, H. Effect
of sotalol on heart rate variability assessed by Holter monitoring in patients with ventricular arrhythmias. Am. J. Cardiol.
72:67A–71A;1993.
Jensen, J.; Axelsson, M.; Holmgren, S. Effects of substance P
and vasoactive intestinal polypeptide on gastrointestinal
blood flow in the Atlantic cod Gadus morhua. J. Exp. Biol.
156:361–363;1991.
Katona, P.G.; Martin, P.J.; Jih, F. Neural control of heart rate:
A conciliation of models. IEEE Trans. Biomed. Eng. 23:164–
166;1976.
Katona, P.G.; McLean, M.; Dighton, D.H.; Guz, A. Sympathetic and parasympathetic cardiac control in athletes and
nonathletes at rest. J. Appl. Physiol. 52:1652–1657;1982.
Keen, J.E.; Aota, S.; Brill, R.W.; Farrell, A.P.; Randall, D.J.
Cholinergic and adrenergic regulation of heart rate and ventral aortic pressure in two species of tropical tunas, Katsuwonus
pelamis and Thunnus albacares. Can. J. Zool. 73:1681–1688;
1995.
Koizumi, K.; Kollai, M.; Terui, N. The physiology of cardiac
innervation: Relationships between cardiac vagal and sympathetic nerve activities. J. Auton. Nerv. Sys. Suppl. 161–171;
1986.
Kollai, M.; Jokkel, G.; Bonyhay, I.; Tomcsanyi, J.; Naszlady,
A. Relation between tonic sympathetic and vagal control of
human sinus node function. J. Auton. Nerv. Syst. 46(3):273–
280;1994.
Levy, M.N.; Martin, P.J. Neural regulation of the heart beat.
Ann. Rev. Physiol. 43:443–453;1981.
Lin, Y.-C.; Horvath, S.M. Autonomic nervous control of cardiac frequency in the exercise-trained rat. J. Appl. Physiol.
33(6):796–799;1972.
Malseed, R.T.; Harrigan, G.S. Textbook of pharmacology and
nursing care. Philadelphia, PA: J.B. Lippincott; 1989.
Meghji, P.; Burnstock, G. The effect of adenyl compounds on
the heart of the dogfish. Scyliorhinus canicula. Comp. Biochem.
Physiol. 77C:225–257;1984.
Potter, E.K.; Ulman, L.G. Neuropeptides in sympathetic
nerves affect vagal regulation of the heart. News Physiol. Sci.
9:174–177;1994.
Rigel, D.F. Effects of neuropeptides on heart rate in dogs:
Cholinergic and Adrenergic Tones in Teleosts
Comparison of VIP, PHI, NPY, CGRP and NT. Am. J. Physiol. 255:H311–H317;1988.
34. Rigel, D.F.; Lipson, D.; Katona, P.G. Excess tachycardia: heart
rate after antimuscarinic agents in conscious dogs. Am. J.
Physiol. 246:H168–H173;1984.
35. Rosenblueth, A.; Simeone, F.A. The interrelations of vagal
and accelerator effects on the cardiac rate. Am. J. Physiol.
110:42–45;1934.
36. Saito, T.; Temma, K. Effects of left and right vagal stimulation
139
on excitation and conduction of the carp heart (Cyprinus carpio). J. Comp. Physiol. 111:39–53;1976.
37. Warner, H.R.; Cox, A. A mathematical model of heart rate
control by sympathetic and vagus efferent information. J.
Appl. Physiol. 17:349;1962.
38. Xiang, H.; Taylor, E.W.; Whitely, N.; Randall, D.J. Modulation of noradrenergic action by neuropeptide Y in dogfish
(Squalus acanthias). Physiol. Zool. 67(1):204–215;1994.